Black & Tan 332iT

[Bry5on]

Randomly stumbled upon a picture of the underside of an F97 X3M and thought of your latest project



Different purposes served by each, but the approach (at least in terms of fastener location on the brace) is similar.

That looks a whole lot like the e46 non-m cabrio

[heinzboehmer]

Randomly stumbled upon a picture of the underside of an F97 X3M and thought of your latest project



Different purposes served by each, but the approach (at least in terms of fastener location on the brace) is similar.

[Bry5on]

Pleased to see the hat still makes a sneaky appearance in this latest post 😋

Really like that approach of tying the v-brace directly to the subframe - will be really interested to hear how it goes.

For some reason I'm struggling a little bit to visualize it - the arms of the bracket that stick down bolt through two new holes in the v-brace right? What I can't quite work out is where the main body of the bracket is bolting to as I didn't think there was as much vertical space there before you ran into the diff (I mean obviously I'm wrong, as you have a 3D scan of it haha). Is it two new holes in the front lower section of the subframe?

Here?

Hahaha that’s hilarious that you spotted the hat. Yes you’ve got the location correct, the plan is to pop two rivet nuts into the subframe there for ease of installation. The drain hole is left clear to prevent any future rusting from the inside out.

Honestly I would prefer a single joggled sheet that bolts through the vbrace fastener, but that would space the vbrace down even further than it already is (damn subframe plates). If this ends up being fiddly to install but the results are conclusively good, i might just weld a simple sheet to the factory U-brace above the v-brace.

I picked up a spare v-brace yesterday from a friend who was sitting on a lot of them, so the worse of the two will be getting holes drilled into it to fit this.

[karter16]

Pleased to see the hat still makes a sneaky appearance in this latest post 😋

Really like that approach of tying the v-brace directly to the subframe - will be really interested to hear how it goes.

For some reason I'm struggling a little bit to visualize it - the arms of the bracket that stick down bolt through two new holes in the v-brace right? What I can't quite work out is where the main body of the bracket is bolting to as I didn't think there was as much vertical space there before you ran into the diff (I mean obviously I'm wrong, as you have a 3D scan of it haha). Is it two new holes in the front lower section of the subframe?

Here?

[Bry5on]

Today I had to load the wagon up completely again, so I got to test how easy it is to remove/install the brace, and I also got to drive it without the brace. Yes, indeed, it does make a very large difference. Still struggling for the right words, but the car is a lot less wiggly with the brace and it genuinely does feel like it's riding on stiffer springs. Secondary motions are significantly reduced in the chassis, motions that I hadn't noticed before but won't be able to ignore again..

I also verified that the storage idea for the brace is a no-go as the brace is too long. Will have to come up with something else:


And since I was so happy with this improvement, I decided to take it one step further. I never did like the location that the v-brace bolts in as the brace/beam it attaches to is placed in pretty significant bending. Since I'm running solid subframe bushings, I can tie the v-brace directly to the subframe, rather than indirectly through the support brace in front of it. Here's a shot at how to do that, we'll see if it nets anything:

[Bry5on]

Ah sorry if I gave the wrong impression. I completely agree that bracing to the RSMs as I have with the RACP brace is almost certainly worse than the approach you've taken as far as torsional rigidity goes. The primary purpose for the brace in my case was because I want to reinforce the RACP points. The improvement to torsional rigidity is a very happy bonus, but I would have definitely thought there's more still on the table in comparison to what you've done (which is why I'm also interested in Heinz's suggestion). All I meant to say was that in my case the RSMs are "locked" in place by the brace in relation to the RACP points, so I was wondering whether what you mentioned about the RSMs possibly moving a little as they load/unload would be a little different in my case compared to your case where the RSMs aren't tied in to your brace right?

Interesting re the wind up loads. That makes sense and would explain the difference we're feeling before/after.

Ah yes I’ve got you now. My car has the Vince bar which would be achieving the similar result, but probably with slightly more flex. It certainly could be that, although I’m skeptical.

I have rogue engineering RSMs and I suspect that they are actually softer than stock ones now that I think of it. Maybe that’s what I’m feeling.

[karter16]

I would venture to say that mounting to the RSM is no better (probably worse?) than mounting to the pickup points that I did for torsional rigidity, so we’re likely feeling the same thing.

Ah sorry if I gave the wrong impression. I completely agree that bracing to the RSMs as I have with the RACP brace is almost certainly worse than the approach you've taken as far as torsional rigidity goes. The primary purpose for the brace in my case was because I want to reinforce the RACP points. The improvement to torsional rigidity is a very happy bonus, but I would have definitely thought there's more still on the table in comparison to what you've done (which is why I'm also interested in Heinz's suggestion). All I meant to say was that in my case the RSMs are "locked" in place by the brace in relation to the RACP points, so I was wondering whether what you mentioned about the RSMs possibly moving a little as they load/unload would be a little different in my case compared to your case where the RSMs aren't tied in to your brace right?

Interesting re the wind up loads. That makes sense and would explain the difference we're feeling before/after.

[Bry5on]

I agree - it's a really hard feeling to describe. I think Heinz articulated it well - there's much less of the feeling of the chassis loading up before the suspension starts working. The brace I'm running ties the RACP points directly into the rear shock mounts, so what I've noticed is that there's basically no give before you feel the suspension working. So possibly it feels slightly different to what you're feeling given the RSMs are fairly immovable with my setup? Definitely need seat time to notice the full effect. I could tell when I first drove it that it had made a difference, but it becomes more and more apparent the more you drive. I thought I'd finally got used to it and then had to go over a speed bump on an angle (I usually take them head on for less jostling around) the other day and couldn't believe how stiff the chassis was.

I would venture to say that mounting to the RSM is no better (probably worse?) than mounting to the pickup points that I did for torsional rigidity, so we’re likely feeling the same thing.

On the wind up loads, I wonder if this is related to the roll center height difference front to rear. When initiating a corner, you have geometric roll stiffness mostly in the rear, where as in steady state you have roll stiffness from your springs, so the front-rear balance does actually change from initial turn in. I wonder if the feeling of the stock car is literally windup and relaxation from the effect of geometric roll stiffness. I guess this is all why they say you must have a high chassis stiffness to roll stiffness ratio! Feels nice.

[heinzboehmer]

Oh, very cool! Had no idea that's how it's built under there.

[karter16]

I'm not entirely sure what the isofix points attach to, but they need to be fairly strong, so I would assume it's more than just the top layer of sheet metal.

I was mostly just listing out all the options I could think of . Packaging is likely what would drive the choice of one over the other in the end.

Gotcha - yeah makes sense.

FYI I'm pretty sure it is just the top layer of sheet. (Edit: thinking about this further it must be the case as it was the hole for the isofix bolt that I put the camera through)

Here's a photo of the mounting points for the seatbelts, they're this captive nut/collar welded to the top layer of sheet metal. I don't have a photo of the isofix points handy, but can't see them being more robustly built than the seatbelt points. I suppose for the seatbelt the load is going to be fairly one-directional perpendicular to the plane of the sheet metal, whereas a brace is dealing with loads in the same plane as the sheet metal.

(The orientation of the photo below is weird, but at the top of the photo is the angled top-layer of the seat base. The nearer captive nut/collar is for the RHS seatbelt receptacle and the further one is for the LHS receptacle and bottom point for the centre seat belt.)

(sorry for hijacking your build thread Bry5on !)

[heinzboehmer]

I'm really the wrong sort of engineer for this stuff, but I would have thought that if you tied in to the front subframe bolts it would render the need to tie into the ISOfix points unnecessary right? I believe that the ISOFix points are attached to only the top layer of sheet metal, so would have thought that the subframe points would be more effective for this? Or am I looking at this the wrong way?

I'm not entirely sure what the isofix points attach to, but they need to be fairly strong, so I would assume it's more than just the top layer of sheet metal.

I was mostly just listing out all the options I could think of . Packaging is likely what would drive the choice of one over the other in the end.

[karter16]

To make it bolt in, you could piggyback off the rear seat top latches, rear seat bottom pivot point, isofix anchors, front subframe bolts (if you have vincebar style reinforcements) and maybe even the rear window motor mount points.

I'm really the wrong sort of engineer for this stuff, but I would have thought that if you tied in to the front subframe bolts it would render the need to tie into the ISOfix points unnecessary right? I believe that the ISOFix points are attached to only the top layer of sheet metal, so would have thought that the subframe points would be more effective for this? Or am I looking at this the wrong way?

[karter16]

Thanks! Yeah it's a very hard feeling to describe in words. Sort of like you can really feel the extra stiffness in the little ripple movements of the rear suspension now. This could honestly be the rear shock mounts displacing a bit before the dampers start doing their job, but it's not something you could feel before. Just lots more 'precise' feeling. All good changes. Did you notice any of this 'ripple' sort of behavior?

I agree - it's a really hard feeling to describe. I think Heinz articulated it well - there's much less of the feeling of the chassis loading up before the suspension starts working. The brace I'm running ties the RACP points directly into the rear shock mounts, so what I've noticed is that there's basically no give before you feel the suspension working. So possibly it feels slightly different to what you're feeling given the RSMs are fairly immovable with my setup? Definitely need seat time to notice the full effect. I could tell when I first drove it that it had made a difference, but it becomes more and more apparent the more you drive. I thought I'd finally got used to it and then had to go over a speed bump on an angle (I usually take them head on for less jostling around) the other day and couldn't believe how stiff the chassis was.

[heinzboehmer]

Good point about using nutserts - that would make this a pretty easy installation with some countersunk fasteners if need be. It would need more bolt points because nutserts don’t have a lot of bearing area and can more easily put sheet metal in bending (=fatigue) but that’s not an issue if you’ve already accepted drilling holes. So this is definitely doable for a coupe. Well, I can always scan a car for anyone who wants a starting point

Actually, if you're permanently modifying the chassis, might as well bond in a long u shaped piece with the attachment points for max bearing area. Would need to do something clever to be able to get the brace out easily (leave enough clearance to lift it up and out of the mounting points?), but sounds doable.

[Bry5on]

I think this is what makes the effect of this type of bracing so hard to put into words. I needed ample seat time in my own car after the Slon wall to really understand all the minutia associated with the change.

Best scenario I've found to highlight the improvement is a sudden and fairly large steering input. Not a jerky move, but moreso something that really loads up the outside tires quickly. Without the bracing, you can feel two very distinct "settlings" after turning the wheel. One being the suspension reacting to the change and the other being the chassis "untwisting" after the intial load. Both of these are still present with the bracing, but there's significantly less time elapsed between the two and the chassis settling is less pronounced. I think this also has the effect of making it feel like the suspension is working harder, like you mentioned, cause that's mostly what you're feeling now. Hard feeling to describe for sure. Much easier to understand in the driver's seat.

Also, I think something similar could be replicated for coupes. To make it bolt in, you could piggyback off the rear seat top latches, rear seat bottom pivot point, isofix anchors, front subframe bolts (if you have vincebar style reinforcements) and maybe even the rear window motor mount points. Nutserts make it not a bolt in mod, but allow for easy pick up points along the side of the opening. Now I kinda wish I hadn't bonded the Slon wall into my car, so that I could design something similar to yours, but for coupes...

Attaching a pic of the coupe opening for inspiration

Good point about using nutserts - that would make this a pretty easy installation with some countersunk fasteners if need be. It would need more bolt points because nutserts don’t have a lot of bearing area and can more easily put sheet metal in bending (=fatigue) but that’s not an issue if you’ve already accepted drilling holes. So this is definitely doable for a coupe. Well, I can always scan a car for anyone who wants a starting point